saukko/Abliterated-Dolphin3.0-R1-Mistral-24B
Abliterated vs dphn/Dolphin3.0-R1-Mistral-24B. Looks like abliteration (refusal removal): only the output projections (down_proj, embed_tokens, o_proj) changed, in 40 of 40 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 1/1, Norms 0/81, MLP 40/120, Attention 40/160, lm_head 0/1
1 derivatives checked
- saukko/Abliterated-Dolphin3.0-R1-Mistral-24B-GGUF: Same weights, converted
1 other repos hold byte-for-byte identical weight files.